A study of power storage configurations in residential communities in multiple scenarios accounting for battery degradation

The peak and valley load differences in residential communities are large, and there is a timing difference between the peak hours of load and photovoltaic power generation. As a result, power storage system deployment can employ photovoltaic output to lessen peak and valley load discrepancies. To make power storage design in residential areas more feasible, the research initially examines the influence of charging and discharging rate and depth on the life of power storage batteries to develop a dynamic loss model for power storage batteries and estimate the operational life of the system; secondly, as the goal function, the largest yearly net revenue is used to optimize the optimization of the power storage system; finally uses particle swarm algorithm to simulate the power storage system in residential communities under different operational scenarios. The results indicate that the residential district power storage system is more effective in the grid-connected scenario and validate the reasonableness and effectiveness of the model used, providing a reference for the planning and construction of residential district power storage systems.


Introduction
As "double carbon" targets are introduced and the county-wide rooftop photovoltaic (PV) policy is implemented [1] , the power system will progressively transition to a further form of power system that incorporates a large share of new energy penetration, leading to a decline in power system stability [2] .Power storage can enhance the capacity of new energy resources, and power storage could "cut the peaks and fill the valleys" for the customer side of the load, decreasing the peak-valley differential [3] .Hence the cooperative operation of power storage and new energy generation will be the new distribution network's development trend [2] .At the same time, the residential load has a large peakvalley differential, and the peak hours of electricity consumption don't coincide with the peak hours of photovoltaic power generation, etc.To address these shortcomings, an Energy Storage System (ESS) can be installed on the transformer secondary side [5] to absorb the PV energy production and reduce the peak-valley differential of the residential cell load.Therefore, before the construction of the ESS in the community, it is crucial to plan and configure it reasonably to obtain higher economic benefits.
To address the problems related to the deployment of power storage for residential customers equipped with photovoltaic cells.By deploying solar and power storage systems for rural microgrids, the article [6] enhanced the utilization rate, economy, and ecological environment of rural regions; the paper [7] has connected power storage to the grid, which has improved the problems caused by the grid connection of photovoltaic and increased the economy of the system; literature [8] configured power storage to maximize monthly integrated returns, but the capacity cannot change every month, so the total economic evaluation of power storage isn't accurate enough, so the whole life cycle of ESS needs to be taken as the time scale for consideration; literature [9] considering the influence of charge or discharge depth and charge or discharge rate on the lifetime of power storage, and reflects the value of secondary use in terms of losses, but doesn't specify the conditions for battery retirement.The above literature mainly focuses on the power supply side, distribution network side and industrial users on the customer side, but doesn't analyze the configuration of power storage in urban residential areas, and doesn't fully consider the factors influencing the life of power storage batteries or doesn't consider the replacement conditions of power storage batteries, resulting in an inaccurate estimation of the service life of power storage batteries.
According to the above analysis, this text examines the effect of charge and discharge timings and depth of charge and discharge on the capacity degradation of power storage batteries in depth.It defines the power storage health state model and life decay model, constituting the dynamic loss model of power storage batteries.According to this model with the highest average annual revenue as the target, through comparison analysis, the calculation example employs the particle swarm method to build a suitable operating scenario and appropriate battery capacity for a residential community and validate the model's logic and efficacy.

Model for power storage capacity loss
The capacitance of power storage batteries decays over their life cycle due to various factors.Therefore, when optimizing power storage configurations, a more refined power storage life loss model is required to determine the benefits of BESS over its service life more accurately [10] .Only the influence of charge/discharge timings and depth of charge/discharge on power storage battery capacitance is studied in this text.
To ensure the power storage battery's consistent output, it is usually specified that when the remaining available capacity decays to 80% of the rated capacity, it denotes that the power storage battery has reached the end of its useful life, and the expression is shown in equation (1)   off R 0 8 S .S (1) In this paper, the state of health (SOH) of a power storage battery is defined using the capacity of the battery as the reference quantity.
Equation ( 2) is the SOH model expression, where: Now S denotes the power storage battery's current available capacity.When 80 SOH % the condition is met, the power storage battery is withdrawn from service and enters ladder recycling.
Based on the experimental data from the literature [11] , the literature [12] carried out a normalization of the data to derive an experimental curve for the connection between power storage SOH and the degree of life loss and fitted the curve to derive a fitted function for the BESS capacity loss as in equation ( 3 (3) To simplify the model calculation, the fitted function is linearized by the following segments, and the simplification steps are as in equations ( 4) and ( 5).O are the linearised power storage decay coefficients.

Model for power storage life decline
Based on the literature [13] , we developed a linearized power storage lifespan degradation model, Eq. ( 6) is the fitted function of the number of cycles of the BESS to the charging and discharging depth; 0 D and 1 D are the fitting coefficients.DOD SOC( t ) ( 7 ) Equation ( 7) is the connection between DOD and state of charge (SOC) of the power storage cell; Equations ( 8) and ( 9) calculate the value of the lifetime loss ( ) Loss t * .
( ) SOC t denotes the cell's SOC for ESS at time t.
3. Model for optimal configuration of power storage systems 3.1.The objective function for optimal power storage allocation In equation (10), F is the ESS's yearly average net return during its whole cycle; E C is the return on electricity sales over the ESS's total life; re C is the is the return from the laddering of the ESS; in f is the initial investment cost of ESS; PV f is the investment expense of PV system; P f is the power expense of ESS; mt f is the O&M cost of the ESS; and N is the service life of the ESS.
In equations ( 12), ( 13) and ( 14), p k , s k and R P denote the expense per unit of power, expense per unit of capacitance and rate power of the ESS.
) In equation (15), γ the recovery factor A is defined by the current market environment [14] .
In equation ( 16), mt k is the yearly cost of operation and maintenance per power unit.
In equation ( 17), PV k is the PV cell's cost per unit capacity; PV S is the rated capacity of the PV system configuration.

Binding conditions
System power balance constraint: are the load powers and working hours that the ESS can operate with backup power, respectively; max R S is the maximum capacity of the BESS that can be configured.

Solution method
Compared with other algorithms, the particle swarm optimization algorithm has the advantages of simplicity, fast convergence and fewer parameters to be adjusted.Still, the standard PSO algorithm has certain shortcomings, making it difficult to take into account both local and global convergence and also easy to fall into local optimality [14] .The solution flow is shown in Figure 1.

Introduction to the cell power system
In this text, we take the PV output and its load of a district in a specific region as an example, the installed capacity of the district is 180kW, according to the PV output under different seasons and weather conditions and the load situation under different seasons, we select the PV historical data under 6 typical weather conditions in the region respectively and 4 types of cell load demand scenarios.Combining the statistical days of various types of weather in one year conditions as a percentage of the year, weighting factors were taken for the PV output and load scenarios under different weather conditions.The weighting factors for the six types of weather scenarios and four types of load demand scenarios show in Table 1 and Table 2 to obtain the weighted combined PV output and load power, as shown in Figure 2. The time-of-day tariff is based on the local tariff system, and the time-of-day tariff prices for each period are shown in Figure 3.In order to verify the effectiveness of this power storage decay model applied to residential district power systems and its impact on the economics of power storage, different scenarios of power storage operation are set up to verify.
Scenario 1: ESS is configured taking into account the influence of power storage battery operation on life loss and capacity decay, with the power storage system using only PV output to supplement the electricity and discharging it during peak load hours, with the optimization objective being the highest average annual net return over the whole life cycle.
Scenario 2: ESS is configured without considering the influence of power storage battery operation on life loss and capacity decay.The rest of the conditions are the same as in scenario 1.
Scenario 3: The ESS is configured to take into account the influence of power storage battery operation on life loss and capacity decay, and the power storage system is linked to the grid so that the power storage operates in the arbitrage mode of "low storage and high incidence", the rest of the conditions are the same as in Scenario 1.

Simulation parameter setting
The "photovoltaic-power storage" system uses a lithium iron phosphate battery as the power storage battery, the battery parameter values refer to the literature [13] , The specific parameters are shown in Table 3; The parameter settings show in Table 4; the power storage SOH and life loss degree relationship curve after fine linearization of the parameter values show in Table 5.Table 3.Values for power storage configuration parameters.
Table 4. SOH segment linearization parameters for power storage batteries.

Analysis of simulation results
To solve the algorithm, use the power storage configuration model mentioned above, Figure 4 depicts the iterative process of the particle swarm method (taking Scenario 1 as an example), and the iterative u .Scenario 1 simulation results show in Table 4.The purchased power of the residential load in Scenario 1 and the operation of the power storage battery are shown in Figure 5.It is clear that, after the arrangement of power storage, the ESS and charging load is supplemented by PV during normal periods, and the stored power is used by the load during peak hours, this has the effect of "peak shaving" and minimizes the load's peak-to-valley discrepancy.The whole life cycle cost of a power storage system / ×10,000 RMB 385.0 The average yearly cost of power storage operation and maintenance / ×10,000 RMB 9.92 The experiment results under scenario 2 are shown in Figure 6 and Table 5. Comparing the simulation results of Scenario 1, the average annual net benefit without considering life loss is RMB 1,205,800, which is RMB 27,200,000 less than when life loss is considered, a reduction of 2.26%; the PV capacity utilized for power storage is 97.4kW, accounting for 54.11% of the installable capacity; and the number of years required for maximum benefit is 25 years, reflecting unreasonableness with actual operation.It shows that considering storage battery losses not only improves the economic efficiency of the storage system and reduces its operation and maintenance costs, but also allows a more accurate prediction of the service life under greater economic efficiency.
The experiment results under scenario 3 are shown in Table 6 and Figure 7.Analysis of the simulation results compared to Scenario 1 shows that the power storage system operating in Scenario 3 not only increases the average annual net revenue by RMB 130,100,000 compared to Scenario 1, but also reduces the capacity configuration by 47%, thus showing that the operating conditions of Scenario 3 not only improve the economic benefits of power storage, but also reduce its footprint size, making it more suitable for deployment in residential communities.However, the PV supplements the power storage system with only 24.04% of the buildable capacity of the plot, causing a large amount of available space in the plot to be wasted when the PV only serves the power storage.
According to the results of the comparative analysis of the three scenarios, the economics and the effect of reducing the peak-to-valley difference are analyzed.Comparing Scenario 1 and Scenario 2, the ROI of considering the dynamic life loss of the power storage is 4.894% higher than the ROI without considering the life loss, and comparing Scenario 1 and Scenario 3, the best effect of reducing the peakto-valley difference is obtained when the power storage is operated in Scenario 3. It is 80.08kW lower than in Scenario 1, and the ROI is also 5.51% higher than in Scenario 1.It leads to the conclusion that power storage systems configured and promoted in residential communities are more suitable for operation in scenario 3.

Conclusion
(1) This paper introduces the PV output under various climatic conditions and the load demand under different seasons, making the configuration results more accurate and practical.At the same time, the composition of power storage for the community can not only obtain economic benefits, but also play the role of power storage system "peak and valley reduction", from the configuration results can be concluded that the configuration of the power storage system capacity is not large, can meet the needs of the small residential area.
(2) In the case of grid-connected power storage, power storage can play the functions of peak-valley arbitrage and "peak-shaving and valley-filling", which has better economic benefits than when it only plays the role of "peak-shaving", and the peak-valley differential of the community load is smaller.
(3) Adopting the dynamic decay model of power storage, the service life of power storage batteries can be predicted more accurately, and the number of times and depth of charging and discharging can be arranged more reasonably to obtain greater economic benefits.

Fig 2 .Fig 3 .
Fig 2. Integrated PV output power and cell load power Photovoltaic power output Cell load power

Table 1 .
Weights for the six weather scenarios.

Table 2 .
Weighting of the four types of power storage operation scenarios.

Table 5 .
Simulation results for Scenario 1

Table 6 .
Simulation results for Scenario 2

Table 7 .
Simulation results for Scenario 3